{
  "id": 234384,
  "title": "inconsistent labeling ",
  "url": "/competitions/hubmap-kidney-segmentation/discussion/234384",
  "author_name": "",
  "post_date": "2021-04-24T02:31:38.331087400Z",
  "votes": 17,
  "comment_count": 7,
  "views": 0,
  "content": "<p>I found glomeruli like the photos below appeared to be true label in some images, while not true label in other images. I understand that the \"level of injury is on a continuous scale\", but it seems that images are labeled by different pathologists with different standard of injury level?</p>\n<p><img src=\"https://i.ibb.co/sFf9zHq/image.png\" alt=\"\"></p>",
  "messages": [
    {
      "id": "1282511",
      "postDate": "04/24/2021 02:31:38",
      "content": "<p>I found glomeruli like the photos below appeared to be true label in some images, while not true label in other images. I understand that the \"level of injury is on a continuous scale\", but it seems that images are labeled by different pathologists with different standard of injury level?</p>\n<p><img src=\"https://i.ibb.co/sFf9zHq/image.png\" alt=\"\"></p>",
      "rawMarkdown": "I found glomeruli like the photos below appeared to be true label in some images, while not true label in other images. I understand that the \"level of injury is on a continuous scale\", but it seems that images are labeled by different pathologists with different standard of injury level?\n\n![](https://i.ibb.co/sFf9zHq/image.png)",
      "votes": null
    },
    {
      "id": "1283176",
      "postDate": "04/24/2021 16:45:31",
      "content": "<p>which image is this?</p>",
      "rawMarkdown": "which image is this?",
      "votes": null
    },
    {
      "id": "1283241",
      "postDate": "04/24/2021 18:09:53",
      "content": "<p>The image is just an example. I found some glomeruli like the image are true on d488c759a, but not true in some other test tiffs. </p>",
      "rawMarkdown": "The image is just an example. I found some glomeruli like the image are true on d488c759a, but not true in some other test tiffs.",
      "votes": null
    },
    {
      "id": "1283679",
      "postDate": "04/25/2021 07:12:55",
      "content": "<p>how to know it's not true in other images. would you mind giving an example to prove it. </p>",
      "rawMarkdown": "how to know it's not true in other images. would you mind giving an example to prove it.",
      "votes": null
    },
    {
      "id": "1284294",
      "postDate": "04/25/2021 19:05:41",
      "content": "<p>You just compare the public score with or without that kind of glomeruli</p>",
      "rawMarkdown": "You just compare the public score with or without that kind of glomeruli",
      "votes": null
    },
    {
      "id": "1284412",
      "postDate": "04/25/2021 22:08:46",
      "content": "<p>yep, I can confirm that one. There is an extensive discussion in an older thread on this. I believe the 57xxxx tiff has some of these dark gloms unmarked. this creates confusion, as one doesn't know how these are treated in the private dataset :(</p>",
      "rawMarkdown": "yep, I can confirm that one. There is an extensive discussion in an older thread on this. I believe the 57xxxx tiff has some of these dark gloms unmarked. this creates confusion, as one doesn't know how these are treated in the private dataset :(",
      "votes": null
    },
    {
      "id": "1284415",
      "postDate": "04/25/2021 22:16:02",
      "content": "<p>Yes, that's what I found when I add the dark gloms to 57XXX. I don't know if the private tiffs count these dark gloms or not..</p>",
      "rawMarkdown": "Yes, that's what I found when I add the dark gloms to 57XXX. I don't know if the private tiffs count these dark gloms or not..",
      "votes": null
    },
    {
      "id": "1284993",
      "postDate": "04/26/2021 13:30:36",
      "content": "<p>thanks for sharing</p>",
      "rawMarkdown": "thanks for sharing",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 1283176,
      "author_name": "andrasferenczi",
      "author_url": "",
      "post_date": "04/24/2021 16:45:31",
      "content": "<p>which image is this?</p>",
      "votes": null,
      "replies": [
        {
          "id": 1283241,
          "author_name": "lililycai",
          "author_url": "",
          "post_date": "04/24/2021 18:09:53",
          "content": "<p>The image is just an example. I found some glomeruli like the image are true on d488c759a, but not true in some other test tiffs. </p>",
          "votes": null,
          "replies": []
        },
        {
          "id": 1284412,
          "author_name": "andrasferenczi",
          "author_url": "",
          "post_date": "04/25/2021 22:08:46",
          "content": "<p>yep, I can confirm that one. There is an extensive discussion in an older thread on this. I believe the 57xxxx tiff has some of these dark gloms unmarked. this creates confusion, as one doesn't know how these are treated in the private dataset :(</p>",
          "votes": null,
          "replies": []
        },
        {
          "id": 1284415,
          "author_name": "lililycai",
          "author_url": "",
          "post_date": "04/25/2021 22:16:02",
          "content": "<p>Yes, that's what I found when I add the dark gloms to 57XXX. I don't know if the private tiffs count these dark gloms or not..</p>",
          "votes": null,
          "replies": []
        }
      ]
    },
    {
      "id": 1283679,
      "author_name": "southsakura",
      "author_url": "",
      "post_date": "04/25/2021 07:12:55",
      "content": "<p>how to know it's not true in other images. would you mind giving an example to prove it. </p>",
      "votes": null,
      "replies": [
        {
          "id": 1284294,
          "author_name": "lililycai",
          "author_url": "",
          "post_date": "04/25/2021 19:05:41",
          "content": "<p>You just compare the public score with or without that kind of glomeruli</p>",
          "votes": null,
          "replies": []
        }
      ]
    },
    {
      "id": 1284993,
      "author_name": "matinroom",
      "author_url": "",
      "post_date": "04/26/2021 13:30:36",
      "content": "<p>thanks for sharing</p>",
      "votes": null,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "1282511": "I found glomeruli like the photos below appeared to be true label in some images, while not true label in other images. I understand that the \"level of injury is on a continuous scale\", but it seems that images are labeled by different pathologists with different standard of injury level?\n\n![](https://i.ibb.co/sFf9zHq/image.png)",
    "1283176": "which image is this?",
    "1283241": "The image is just an example. I found some glomeruli like the image are true on d488c759a, but not true in some other test tiffs.",
    "1283679": "how to know it's not true in other images. would you mind giving an example to prove it.",
    "1284294": "You just compare the public score with or without that kind of glomeruli",
    "1284412": "yep, I can confirm that one. There is an extensive discussion in an older thread on this. I believe the 57xxxx tiff has some of these dark gloms unmarked. this creates confusion, as one doesn't know how these are treated in the private dataset :(",
    "1284415": "Yes, that's what I found when I add the dark gloms to 57XXX. I don't know if the private tiffs count these dark gloms or not..",
    "1284993": "thanks for sharing"
  },
  "source": "meta"
}